Binpress Podcast Episode 36: Jonathan Deutsch of Tumult

The Binpress Podcast

This week we talk with Jonathan Deutsch, founder of Tumult, the software outfit behind Hype. If you’re not familiar with Hype, it’s a Mac app that helps users create stunning animated and interactive HTML5 content, and it’s used by both beginners and pros.

Jonathan covers how he got his start, the importance of a support network, how your product itself can drive marketing, why you should use a public support forum, and much more.

Listen to the podcast in the player above, or click here to download it directly. Subscribe on iTunes or do so manually by using this RSS feed.

Subscribe oniTunesPocketCastsStitcher

Show notes

• Jonathan Deutsch: Website, GitHub, Twitter
• Tumult
• Hype: Website, Twitter
• HypePro
• HyperEdit

Transcript

Alexis: Jonathan, thanks for coming on the podcast.

Jonathan: Glad to be here.

Alexis: So before we get to Tumult and Hype, tell us a little bit about your background.

Jonathan: Sure. Before Tumult, I started in CS at Purdue University and right out of college I was at Apple and have worked some various jobs at Apple for about six and a half years. I always wanted to start my own company; Apple was a little bit of a detour and so the opportunity came up in the HTML5 space. I had been there long enough that I felt I had accomplished what I wanted to do, and so I left and started Tumult.

Alexis: Alright. Let’s take another step back here. How did you learn how to program? What got you interested?

Jonathan: I would say having a TI-82 calculator was probably the first thing that really sparked my interest in programming. They were programmable calculators; you could do BASIC programming on it. You could also do Assembly programming, which I didn’t do much of, but I thought it was so much fun that I could write a little program – and usually they were text adventure games, is what I really started doing.

I did Escape from West Middle – West Middle was the middle school I was at, and so you could choose one of different routes. It was very much like, if the user chose this, go that way; if the user chose that, go this way. It’s like you would have to start in your classroom and then eventually you’d escape out to the hallway. You’d have to avoid the administration and try to get your way out.

That’s what started me; it was really just more fun and games and if I could impress my friends and I’d give them my calculator –.

Alexis: I’ve got to say though, I did daydream about escaping from middle school or high school, but putting it into a text adventure is taking it a step further [chuckles].

Jonathan: Yeah, I think it’s every kid’s dream. In fact, I would do it in BASIC, but I also did Hypercard versions of it as well. We had like a QuickTake 100 camera so I’d take photos around the school, and then each card would be a photo so you could choose to go right, go left – things of that nature – and you could digitize videos, so I’d have you interacting with various students along the way. I had this recurring theme and all the projects I did was trying to get out of middle school.

Alexis: [Chuckles] But man, that sounds like a calculator project that would have been very simple, but you even had a full-motion video and –.

Jonathan: Well, the Hypercard version had video.

Alexis: The Hypercard version, okay.

Jonathan: The calculator ones, you couldn’t do a whole heck of a lot with – but that at least sparked my interest. I had been into – for Hypercard, I was more into video projects. I like doing video editing as well, so I was really doing more of the multimedia stuff side of using computers at the time.

Through sheer luck, I happened to get a job at a dotcom company, and they were a search engine. You would enter in audio and video manually into the database so that users could search for streaming audio and video.

And so I did this little project to help enter links in faster and the next day everyone was using my little JavaScript project. When I saw everyone using a program I had actually created to make their lives better – in fact, the faster you went the more they got paid, so it kind of helped them make more money. At that point I knew, programming is what I wanted to do and I was hooked at that point.

Alexis: And when was this? In middle school still or high school?

Jonathan: By the time I did that, that was in high school. That was, I think, about junior year is when I took the leap from “Okay, I want to use computers and do video effects –.” No, it was actually the programming aspect and writing those applications, not using the applications. Writing the applications is what I wanted to do.

Alexis: So you dived into college, straight into CS, right?

Jonathan: Yeah.

Alexis: So you studied computer science, and then you’ve applied at Apple, and lo and behold [chuckles] you go into the white, pearly Apple gates.

Jonathan: Yeah, it was funny because in college, I always figured “Well, I’ll start my own company.” I always felt like I had an independent streak; I was always making side and small businesses.

And then a friend applied to an internship at Apple, and I was a pretty big Apple fan boy. I was watching all the Steve Jobs keynotes at this point; I had a first gen iPod. It was funny, but I never actually considered working at Apple until he applied.

At the time, Apple was really much more of a hardware company; they didn’t have a lot of software. Especially back in the ‘90s, it was about the operating system.

But you look at what software you use today – Safari was not an application at the time. iChat, or what’s now Messages, was not an application at the time. The Calendar was not an application; iTunes came later.

My mental image of Apple was much more of a hardware company, and so it never really occurred to me that I should apply to this company that I love until I did a second glance when a friend got the job and realized, “Oh, they do a lot of application development and they do a lot of things I’m interested in.”

So I decided, as I was graduating, I applied to some various companies to cover my bases –.

Alexis: Just to do your due diligence, so to speak. “I did the responsible thing.”

Jonathan: Yeah, and then when I got the job offer at Apple I’ve just realized there’s no way I could turn it down.

Alexis: So what did you wind up doing at Apple?

Jonathan: My first job was on a team called Software Update and Integration. You could hear that name and be like, “What the heck do they do?”

Alexis: I picture OSX updates would pop up and – something like that. Is that along the lines of what you were doing?

Jonathan: It is along those lines. However, we weren’t the ones in charge of the software update application itself; that was another team. The software updates, the content in the updates, all came from various teams within Apple, so we weren’t the ones actually writing the code that’s shipped in the updates.

We were the ones more or less making sure that the updates all worked together, so it was very much a glorified QA team. By the time I got there, there was less manual QA and the team was really shifting a lot more towards automation and automated testing. So most of what I did on that team was writing internal automation tools, things that would drive the mouse and keyboard and test, or do lower-level testing of making sure the file system was intact, or making sure networking worked.

I was responsible for this – there was an update, I think it was 10.2.8, where for some class of users, networking was completely broken. And so Apple shipped out this update and then had no recourse, no way to ship out new updates because their networking was broken, so that was a huge fiasco. It was now the team’s mission to make sure that type of thing would never happen again.

I think the first update I worked on was 10.3.4 or 10.3.5, so it was back in the Panther days. We did a lot of automated testing and eventually I moved to manage that team. We did way more automated testing and worked on tools that were used to route Apple to write and share automated tests and to drive those tests and report back results and kind of get the whole holy grail flow of integration and automated testing.

Alexis: I’m curious – you mentioned earlier that you had that itch, the entrepreneurial tendencies to just do something on your own. Did you have to suppress that itch while at Apple, or were you always looking for a project. “Maybe this is what I’m going to build” or did it just come to you like “Okay, wow. This HTML5 space is interesting and this is where I need to stake my claim”?

Jonathan: The first team, which was the Software Update and Integration team, especially when I started, we had a lot of autonomy. We were able to do a lot of our own projects and so I was able to start projects that I thought were personally interesting. I was able to lead those projects that I thought were interesting later down the road, and we didn’t have a lot of oversight from upper management because we were a little bit off to the side.

We weren’t the ones actually doing the products and we were doing things internally, so we weren’t really in the spotlight and there weren’t a lot of shareholders.

Alexis: You were the pirates within the Apple navy.

Jonathan: [Chuckles] Actually, one director called us cowboys, which was probably an accurate description. Sometimes we’d piss off a couple other teams because we’d find a lot of issues, things of that nature.

So that team was pretty good. But then later I got the idea, “Well, I need to try something else. I’d like to do a little bit more product engineering at Apple.” So I moved and I was managing the core components of Mail for OSX.

That team was a little bit different. I loved the engineering team, but there were a lot of shareholders for Mail – everyone used Mail, including all of upper management, all the users, and so it was very difficult to do what you wanted with Mail.

There’s a lot of good, legitimate reasons for that. I think when you’re working on a project that’s used by tens of millions of people, you don’t want to move too fast and break things because you’re going to break their workflows, and that’s completely understandable. But it really did go a little bit against the grain of my personality and I think I realized that now.

So even though I enjoyed my time on that team, I think there was a little bit of the “I like to be 100% in-charge of my own destiny.” At that time, it was a matter of the timing being right that I wasn’t really feeling like I was able to use the entrepreneurial aspects of my personality, and this whole HTML5 thing came up.

Although in reality, it was more of the downfall of Flash than necessarily the ascendence of HTML5. Those factors conspired to make me think there’s an opportunity in that space. I had been there longer than I ever expected because I did truly enjoy working at Apple. It was fun; there were a lot of smart engineers there.

Alexis: A kind of transitional question here that covers both your time before Apple, your time before Apple, your time at Apple, and now your time at Tumult. Going from a single programmer who’s working on their own project to managing teams – not necessarily a whole different but new skill sets that have to come into play there, or new muscles that you have to flex – how has that changed for you?

Jonathan: I think there’s a few different ways of looking at the world. I think when I started out making the transition from being just an engineer to an engineering manager, I tried to think of people almost as an engineering problem that we had certain tasks and people’s certain skills, and could you fit all the pieces together. I think that’s a simple and ultimately a relatively effective approach, but I think it misses a lot of the nuance of management.

I don’t know if there’s a good answer to the skill set required, but I’ve led more people – yeah I don’t know if I have a good answer on that [chuckling].

Alexis: You just got to get used to it.

Jonathan: Yeah. Fundamentally, there are very different aspects. I think I’ve been extremely blessed to work with some really talented people who also are quite independent and don’t need a lot of oversight. I think when a lot of people think about management, they think about all the interpersonal problems that people have and how difficult the job of hiring and firing can be.

For the most part, I haven’t had to deal with a lot of those issues, so I feel I’ve just gotten really lucky. I think some of it also comes about in how you structure and how you lead a project. Ultimately, I a lot of project management decisions can also need to personality conflicts that don’t necessarily need to ever happen if you structure a project, even from an engineering level, very well.

If you have loosely coupled components where two people who might fight like cats and dogs have their own separate components, they’re not necessarily going to interact and can still be extremely productive. So I think as a manager you have to not just look at the people and realize there’s going to be inherent conflicts, but you can also look at what is the architecture of this program and how does this architecture actually influence how people will interact in positive or negative ways.

You may want places where there are tightly coupled components; have those people who really can get along and have shared values in their development style work together as well. I mean, there can be two people who are great friends but develop in completely different styles and have completely different thoughts on development and then they’ll end up clashing. So you have to look at various different aspects of people and figure out where that fits together and how you can even structure the program at an architectural level so that people can work very smoothly on it.

Alexis: So when you started Tumult, how big was the team there?

Jonathan: It was just two people – it was me and another guy whom I have worked with and was friends with at Apple.

Alexis: Since we’re all kind of in the hiring track mindset, what do you look for when you need to expand the team and what were some of your first hires?

Jonathan: Our very first hire was a guy by the name of Daniel who – we got really lucky with him, but we needed someone to come in and help with support.

We launched version 1.0. The point of Hype was to let graphic designers build animated websites in a similar manner to how they would do Flash-based content. Flash doesn’t work on the iPhone and iPad, so you need to use a different technology, and that technology that Apple’s really supporting and developing was HTML5. You could run HTML5 on the iPhone, the iPad and it had a lot of great features as far as being able to manipulate quickly for animations, had video capabilities, upcoming audio capabilities, so it seemed like HTML5 would be the future.

So when we released, we got a lot of press because we’re really the first real tool and real 1.0 tool that did this type of HTML5 animation. With all that press, we actually got a good number of users out of the gate, but we were an immature tool. We actually barely did anything at all. We could do animations, we could work with a bunch of different browsers, but there was so much about the web that you don’t know until you’re in the thick of things, like how the text and codings work with different servers, how file names can interact on web servers as well, all the various configuration options with web hosts, how screwed you can be by Chrome doing an update that introduces a bug with 3D transitions that you thought were just a neat feature to have and now suddenly you need to support it across tens of thousands of users.

So we were in this state where we had an immature tool; we didn’t have a lot of knowledge yet about various edge cases and web technologies that we needed to work with, and a ton of users that we need to respond to either with questions about just buying the product or with all these configuration issues.

We were in this horrible hole of customer support where easily we’d have 400 emails unanswered, and the hole was getting worse and worse and worse and sometimes it would be like 20 or 30 days sometimes before we’d be able to get back to certain people. And so we thought we were absolutely failing at customer support, we need to hire someone right away.

And we recognized this. We recognized this pretty early on. “Okay, we need to hire someone in customer support.” But all these problems that made Hype a very difficult product to work with also were issues in hiring people. You needed a very strong level of technical ability to support Hype; it wasn’t just like a “Hey what’s my serial number” and then you give it to them. It was on IE8 in the IE7 compatibility mode –.

Alexis: And only on Windows 7.

Jonathan: Yeah, and only on this version of Windows, with this particular Hype document there’s an issue. It takes a pretty strong level of technical ability to do that, so we needed to hire someone who 1.) could interact with customers really well because that’s what you need in customer support and also had a high level of technical ability, or had done web development before. So that was what we were looking for in our first hire.

We were in this big hole, we need someone fast, and all the things we had tried or people we had tried at that point were just really not working out for one reason or another.

Alexis: Now when it came to expanding the team in terms of development, what were some of the considerations you made there? Because a lot of times I find that folks who have this entrepreneurial streak or this streak of independence “I can do it all myself for mostly all myself,” they get to a point where “Okay, maybe I can’t” and then they realize, “I need help with this other thing that I can normally do very well but I just can’t clone myself or there aren’t enough hours in the day.”

Jonathan: You know, it’s funny. I think we had two really good things going for us at the time that we decided to hire from a development perspective. The first was that – we were in this hole and so myself and my co-founder were doing all the customer support, and so we kind of knew what we needed from a development standpoint done.

We had a really good base of people to pull from that we had worked with at Apple or that we knew from other places, so we were in a pretty good standing to know –. I was a manager and my co-founder was a project manager – we kind of knew how the pieces would fit together when we would start hiring. We had the idea in our head of what software development at the company would be like, and we have this great pool of people to access.

We started out – because we were in this customer support hole – we said, “Well we’re spending so much time on customer support; we could hire a couple of friends we knew to do contract work as development to make sure that at least development is still churning on the product.”

We also contracted out two friends who were extremely talented. At the time we were doing this we were also thinking who should we hire long-term, and so we took a very pragmatic approach. We made the spreadsheet and on the left row we had all the various names of people that we thought were the ones we liked and thought were smart and wanted to work with. And then on the top columns we did all the various attributes that we thought would be important for an engineering position at our company. Some were on the technical level such as “Do you know Objective-C?” because we’re a Mac app. “Do you know JavaScript?” because we also have to deal with the web and web components.

We looked at some future skills we thought might be necessary such as iOS development, which came into play a little bit. We have Hype Reflect, which is a companion app to do previews. But at the time I think the iPad had just come out, we were thinking, “Should we go the iPad app route?”

So we did some technical skills and then we also did some personal skills. I think we had one that was like, “Is this person an asshole” factor.

We just said, “Okay, we have too many people. Let’s take a very scientific approach to hiring because clearly–.”

Alexis: Let’s see – filter by assholes and – [chuckling].

Jonathan: Exactly! So we ranked people, and then we came out with a composite number. We wanted to go one by one; we didn’t want to really extend offers out to more people than we could hire and get their hopes up and then say “Oh no, actually this person filled the role.”

We wanted to take a pretty good approach, and it was pretty clear who was at the top of the list. Those people pretty much became our contractors. One of the people that we contracted out with was someone I had worked on the Mail team with. He just left Apple and moved to Michigan and eventually he was doing a really great job helping us get features done while we were managing some of the business aspects and we weren’t able to develop as much as we wanted.

Eventually I said, “Hey can we pull you back to the Bay Area?” The timing worked out for him so we were able to get him to come back full-time.

Alexis: One thing I’m interested in is your experience with the web before Hype, mostly objective-C background I would imagine. Did you have any experience with web development or a lot of experience? What would you rate your comfort with web development before Hype?

Jonathan: Well on a scale of 1 to 10 I’d probably put myself at about a 6, but that’s relative to a million different skills – at the time.

Alexis: Yeah, it is.

Jonathan: Now I’d rate myself – I don’t know, I’d probably rate myself a 3 given all my [chuckles].

Alexis: All that you know you don’t know.

Jonathan: Exactly. No, but I always had one foot in the app development world and pretty much throughout my entire career as a developer I also had one foot in the web world.

In one of my internships at Purdue, I was working on Serious, which was security center, and we were doing this portal that was in PHP so I was doing a lot with the web there. I worked on another product that was a Mac app called HyperEdit, and HyperEdit was probably the first of its class to do this, although I don’t think I shipped it.

At WWDC, Apple just announced the Safari SDK, which would let you embed a webView into any Macintosh application, so it didn’t have to be limited to a web browser.

And so I put a textView right next to a webView and said, “How fast is Safari?” so you would start typing. I turned that into a product, so that was kind of a mixed product as well that also a webView.

And then at Apple, a lot of our internal reporting tools were all web-based, so I had done a fair amount with the web and web technologies. I have my general head in this space, but I hadn’t done really large amounts of professional development that was used by tons of people. I hadn’t done a lot of JavaScript; I had definitely done some JavaScript and understood the fundamentals.

It wasn’t a space that was completely unknown, and at the same time it wasn’t a space that we were experts in. Definitely we’re more of an expert on the Mac and application development side than the JavaScript side – and that showed in version 1.0, definitely. And we’ve improved our code size, our code footprint, how we deal with things and we’ve learned a lot over the course of the last four years.

Alexis: Jumping back for a second – did you work on Hype before you left Apple and then you’re just like, “Okay, I’m comfortable with where it’s at. Now I’m going to pursue it full time”? Or did you just say, “Okay, I’m done. I am dedicating all my time to building this thing.”

Jonathan: You can’t really do too much work while you’re still at a company, because there are some gray laws on intellectual property. I had done a little bit of experimentation on my own time with my own equipment to see, “Is this feasible?” You wouldn’t want to jump ship.

Alexis: This is just a pipe dream.

Jonathan: Yeah, exactly. I had done enough experimentation to know that something like this would happen, but the real work didn’t begin until we left.

Alexis: Alright, so you mentioned that the press was pretty – how should I say this? It was pretty excited about Hype because as you said, this was a pretty new tool. I mean, nobody had really done this before; it made lives easier. In fact, it made my life easier because I bought it and I said, “Oh man, now I can make some sleek animations without having to bang my head against the wall for weeks upon weeks and make sure that everything looks good.”

Did you plan the marketing and the press outreach way in advance? How did that look for you?

Jonathan: We didn’t plan too much and I think that’s one area we’ve learned a little bit over time. But one thing we did do as a company was we joined a startup incubator called Y Combinator, and we knew that they have some good press contact, so we leveraged that.

One of the big reasons we joined was because they knew press and every company that launched would basically more or less be guaranteed of getting press.

We got an article in Mashable, which was a huge driver of traffic to us. I don’t know how prominent they are today, if people still read them, but back four years ago, they were pretty widely read. That also spurred a lot of secondary press and international press.

At the time, I don’t know if we really knew John Gruber at the time, but we know him now. I think someone said to him, “Hey, there’s this Hype too” and he fortunately linked to us and drove a lot of press our way.

Alexis: And the Apple digerati blessed Hype.

Jonathan: Yeah. And he has such a great readership, so many people in and out of the Mac community follow him and listen to what he has to say, and so that also helped pretty significantly. Those were our two little sparks that lit the fire as far as the press we got.

Marketing has always been a little tricky for us, because Hype as a tool is a very generic app. It starts with a blank slate and you can create anything.

The most common question we get when we talk about marketing is “Who is your most common user?” To this day, we haven’t been able to identify it because people make all kinds of different things from advertisements, to children’s books, infographics; they do digital magazines, and so it’s really hard for us to really focus on one segment and say that is the segment we’re going to focus on.

We’re always in this marketing conundrum of “Okay, maybe this release will try spinning our story this way and see if this is the right market.”

We’ve tried a lot of different things and still to this day, when we do a big release, we get a pretty decent amount of press. And so there’s so much noise – or I shouldn’t say noise – but there is such a flood of new users that drowns out – . Were successful in marketing to this segment? Because we still get a wide variety of users and it’s hard to say – it’s one thing we’re dealing with right now is, are we being successful in certain segments and are we moving into segments where there is money to be had.

Alexis: How do you make sure you get in front of these different segments that you’re thinking about, whether or not they are your audience?

Jonathan: The good news is a lot of people will come to us with opportunities, and so our strategy right now – or should I say lack of strategy – is that, based on our inbound requests, we’ll follow up with those.

One thing we did in Hype 1.5 was we added the ability to export as an iBooks Author widget. iBooks Author was an app that Apple made to help you make digital textbooks, and the one thing we learned was everyone wants to be an author, so this app became extremely popular in the Mac community.

This was early in the days of the iPad, so there was a lot of excitement about what the iPad could do and what digital books would be on a device like the iPad. Between the excitement of the iPad and just everyone wanting to be an author, we decided we’d do this export so you could export your animations to iBooks.

If you were doing a textbook, you could do graphics maybe describing physics equations. Or if you were doing a children’s book, you could clearly do educational content.

And so it was a very small feature; it literally took us about ten minutes to engineer. So we shipped Hype 1.5 with this feature and Hype 1.5 had 30 other major features – this was one of them. Apple loved that we did this, so we got featured by Apple, and on the Feature in the Mac App Store we got the top banner. It had the name Hype, but it also had a little bit of text that said something to the effect of “Can now produce iBooks Author widgets” or something like that.

And so we got a huge influx of people interested in that, and so to this day we get a lot of inbound people who are using Hype with iBooks Author.

One of the things that we’re doing marketing-wise is there is an iBooks Author conference; I think it’s in Nashville. I believe it’s in October – I’m not looking at the dates right in front of me – and so the organizer of that just reached out to us and said, “Hey, I love your product. I think it’s really great for building iBooks. How would you like to come and help sponsor and show off the product as well?”

We said, “That’s a great opportunity” because definitely we have a big segment of our market doing that, and as we said, everyone wants to be an author, so we think there is an opportunity there.

So it was kind of like this little feature we added led to a lot of inbound interest, creating a market and giving us other market opportunities that snowballed from there is the general approach we’re taking right now. There’s so much to do with the business so the people who are excited about us will more or less come to us right now and we’re really happy to engage with them.

Alexis: There are two sides of this – the people coming to you, but also the feature or product-driven marketing side. That happens in other ways in addition to the way it happens with the iBook example, or is that more of an isolated kind of incident?

Jonathan: We definitely look, when we do a new version, at a few different factors. We want to look at where we can put this product and what we can do with the product to open up new markets or to expand new markets, and that’s a big aspect of what we do. And we look for the areas where current users want specific features and those features would also put us in really great positions.

One example of a feature like that is for Hype Pro. We did a feature that lets you do responsive layouts.

Responsive is of course a really big deal with the web right now. The idea is that you can make a single website that will look great on a device of any screen size, going from the iPhone up to the iPad to desktop computers.

You could say, “Websites would always change size,” but when you deal with different device classes like the iPhone or the iPad, sometimes you need different hit target sizes, or the number of columns you have on a page really should be rearranged. It was pretty clear this was a problem for current users, and we had a small solution in place in our previous release, but it wasn’t really cutting it for completely different device classes, and so we got that feedback quite a bit from users.

But we also saw that in advertising – this is also a big deal, that people want to make responsive ads so that instead of doing all these different ads completely separately, you could have one document that would respond to the different device classes. Or if you click the ad, the ad would – I think they call them “expandables,” like open up.

So we saw, “Okay, well this is what our current users want,” and there’s this untapped market for responsive ads. Of course, people in the ad business, those are people who also have money; they’re looking for solutions and there are interesting business opportunities in different business models and just selling single licenses to them.

We said, “Well, we’re not really an advertising company per se, but there is a business opportunity there if we do this feature that will hit our existing users’ needs for doing responsive websites and what could be a new market for us, which would be responsive ads.”

We try to find these feature areas where it’s just like all the different gears are lining up just right and hopefully it will be a feature that helps us out.

Alexis: I’ve got to say, of all the folks that we’ve had on the podcast, I think that you and Hype – the strategy of looking towards the edges and where can we expand, and kind of not necessarily making a whole new product but just adding a certain feature that will open up a new market –. I don’t know if it’s unique to Hype, but at least putting that succinctly, is certainly something that [chuckle] I haven’t gotten when talking to anybody else so far.

Jonathan: Well I think one of the benefits is that Hype itself was always meant to be a cutting edge product. We took a look at HTML5, and the name Hype even comes from many different factors, but one was all these articles were coming out saying, “Is HTML5 living up to the Hype?”

And so it’s very clear that there is going to be this wave of HTML5 and we said, “Let’s ride this wave because it’s going to be growing, there is going to be a whole market behind it. There is going to be new technologies coming out all the time.”

Over the lifetime of Hype, we’ve seen things go from ideas to actual standards, whether it’s the web audio APIs or web GL, which we don’t really take advantage of now – all kinds of new CSS3 features. And so we said, we can kind of ride this wave and there’s going to be all kinds of fallout from that, and all kinds of new technologies that we’ll want to adopt, and business model as well will come out naturally from that. When you have new technologies, you also have new businesses.

Alexis: In doing my research, I did not realize that you were in Y Combinator. This opens up a whole another set of questions. [Chuckles] One of them, at least the one that’s at the forefront of my mind, is what you envision or what you had thought of Hype and Tumult – what size you had envisioned your company being?

Because usually when I think of a company that wants to take an investment, they think “We want to make millions and millions of dollars, we want to be huge, maybe even cross the bridge over into a billion.” When it comes to desktop software, it’s not something that I always think about, something like Y Combinator.

In addition to the context you said you wanted to develop, what was the impetus or the thought process that made you think, “You know, this Y Combinator thing in taking investment is a good idea”?

Jonathan: So there were a few factors, and I’ll get to the investment aspect of it. But the first factor and the biggest factor was that at Apple, we had a very large support network. When you’re in Apple, you’re not working alone even if most of your job – maybe sitting in front of a computer and working alone – that you have tons of other people in a whole company infrastructure to help you out.

And so when I left Apple, I realized all that would be gone. And so I wanted as much support structure as possible to help the business, and it seemed like Y Combinator would be great in that respect because Y Combinator clearly had the partners who had experience starting a company.

They have all kinds of connections that could help us. And then the big thing was really just the other companies we’d be in there with, we’d have this port network and we wouldn’t be doing it alone. We’d have other people both from a – well, their business might help my business, but really more from just a social aspect of it that we have other people we can talk to, other people we can bounce ideas off to. Just a new set of people who would be interesting, honestly.

The largest factor was not wanting to do this alone, and not wanting it to be just me and my co-founder at the time and really getting as much advice and know-hows as we could get.

Now the investment aspect is something entirely different. And we’ve gone a pretty different route than most YC companies. At the time – let’s rewind, which was now nearly five years back – Y Combinator wasn’t as well-known as it was back then. The number of companies was a lot smaller, and my perspective on Y Combinator before going into it was it was more about how you build a company than it was necessarily about how to get further investment rounds.

Y Combinator clearly gives you a little bit of money but it’s quite insignificant, especially for the percentage of the company they take out. Y Combinator’s benefits are all the other things, but the focus on Y Combinator when we got in was significantly more on getting further investment realms. And that, to us, was less important.

You’re right, we are a desktop application; we’re very old school, and that we sell software and people buy it and pay us some money. And that’s kind of the end of it. We don’t give our software away for free in the hope of some later play.

And so our business model wasn’t really conducive to wanting investment because if we did well, we would be making money; that was the goal.

So clearly, we want to grow, we want to be as big as possible. And we thought we were hitting this large wave where maybe we could be as big as Adobe, was what we were saying. We wanted to be the Adobe of the Web, was kind of our initial pitch.

To us though, the independence in having our own control was a bit more important than taking investments that we weren’t sure what we would actually do with the money anyways because we would have some coming in. So it was like I’d rather grow slowly and steadily and make sure we make the right choices along the way, than get a bunch of money and burn out.

Alexis: Right.

Jonathan: So that was kind of the reasoning there. So Y Combinator, in some ways, was not a perfect fit but it was one that still worked for us versus against us, and it was ultimately a good program.

Alexis: Say, if you were building Tumult today, would you still go into Y Combinator?

Jonathan: It’s funny because I think today’s environment is a little bit different and there are things I know now that might make me favor taking investment, in fact.

One of the big ones nowadays is that we are based in San Francisco – I personally love the Bay Area, and the Bay Area is really hot with venture capital money. And so what that does is that inflates crisis tremendously. It inflates engineers’ salaries tremendously. And so for an indie company making software where we’re trying to be very careful with our money, the environment’s not that great.

I just read an article and I think the median price for an apartment in San Francisco is $4,200 a month.

Alexis: Yeah. If it wasn’t that expensive I would live out there. But, no thanks. It’s as beautiful, but not that beautiful.

Jonathan: Yeah. So it’s kind of ridiculous and so much venture capitals flooding in that at this point, you almost need to take some to compete on salaries and benefits. So I think, in my head, some of that competition and some of our struggles could be alleviated by taking more funding now. YC is a great program for finding that funding.

So given what I know now, I would probably – I think it would be good to make changes one way or another. It might’ve been better to never have done YC, but I think I still value the advice we have gotten, the connections, the people we met. There was a lot of value so I don’t regret doing it. I would probably do it again, but if I were to do it again, there’s a high chance I would actually go more with the traditional route and really try much harder, or at least push getting investment because that might be the right choice nowadays. Back then, it didn’t seem like it was, so we didn’t really push it. We kind of flirted with the idea and decided that it wasn’t for us.

Alexis: Now do you still want to be the Adobe of the Web? [Chuckles] Or do you have less grand admissions?

Jonathan: That’s still our goal. I think after being in business for four years, I think we have a better perspective on application development and how our current model hasn’t let us grow to be quite that size as quickly, so I think we have ideas on how, at this point, we can grow our business further.

But I think just that you’ll never be that big by just putting an app out there and hoping that enough people download it. There is not that big of a market, and the way pricing is nowadays, you can’t charge that much and expect a large number of people to pay $600 or $700 for an application.

Alexis: Oh, man. I know you can’t reveal anything yet, but man, major foreshadowing [chuckles] of big things to come.

Let’s go back to the thread of finding your audience or different kinds of audiences that Hype has. You were able to kind of tease out two separate kinds of audiences because you’ve got Hype, and then you’ve got Hype Pro. How did you come to that decision and what has it meant for the bottom line?

Jonathan: For the bottom line, it’s been pretty good. We’ve liked doing Hype and Hype Pro.

Hype Pro technically is an in-app update, so you buy Hype and then you can upgrade. Hype by itself, the standard edition is $50. You can upgrade and get, I think, it’s about a dozen other features for another $50. Or you can – if you buy from us directly, you can do a one-time purchase for $100 and get the whole thing.

And so that came about a little over a year ago. We had just wrapped up Hype 2.5 and we said, “What’s next for our company? What could we do at a product level that will help earn the business? And what can we do that, again, that would propel us perhaps into different markets?”

And so we flirted with a few different ideas. We said, one, we could simply continue working on Hype as it is. Keep adding features, and that could be one thing we could do as a business, and just do a paid upgrade model as we have done between Hype 1.0 and 2.0.” We said we could do an entirely different product not related to Hype at all. Or, we said, we could maybe do a Hype professional model that have these features that could set us up in the future for then using the Hype core technology, because we had invested a lot in our in – like the JavaScript Runtime, and just a lot in the app itself. We had made some good investments there that couldn’t make us leverage this and propel us to the future.

And so we kind of went that third route where we decided, “Okay, we’ll create this feature set that might be of a more limited audience,” but was very important to that audience. It’s one feature which was the notion of symbols, which was something in Adobe Flash, and I hated the name [chuckles] symbols. We had so much debate on the name, and Adobe – you’d always say like, right click and say ‘Convert to Symbol’ and it’s like –.

Alexis: Yeah, I know.

Jonathan: Yeah, it’s like, what does that even mean? That’s like a horrible user experience. We eventually settled on symbols – that’s a whole another story, but there are all these different features that we thought could propel us because we can take this notion of reusable content. And then at that time I think we were of the mind that maybe we could do separate products based off of the reusable content where we could package it a little bit differently and make it for a specific market.

Alexis: Single tools, or –.

Jonathan: Yeah, but ultimately we decided we could also do – with Hype Pro we could do price differentiation so we could kind of capture different market segments that if your needs were more limited, or you like where Hype was at –. Because we didn’t remove features for Hype Pro; we only added features that you could continue using Hype 3.0.

The Hype 1.0 to 2.0 paid upgrade was a little bit of a pain, so we didn’t quite like that model. And if your needs were more sophisticated, you wanted these features, then you could pay for it. So we would have kind of the premium product as well.

It has actually worked out very well for us. First of all, I think we added a lot of value in Hype Pro, and so people have been actually pretty happy to say, “You know what, the product is worth it. These are the features I want.” And so Hype Pro is a very mature product now so people have made that upgrade – there have been a lot of people upgrading.

One of the decisions that helped us make it was we watched a talk by Omni, and Omni also has the standard and professional versions of their products. They said 80% of all their customers pretty much are a go for the Pro version, and we see very similar number to that as well. So that’s surprising and also very good that it’s a higher price point product, and people tend towards that version. So that helps us as a company from an upgrade model. Anyone who had Hype 2.0 would get Hype 3.0 for free, and so they can just buy Hype Pro as what their upgrade would be for that version.

So it wasn’t like people kind of lost the product or lost the track. They just would get Hype 3.0 which was a fabulous update by itself, or at least that’s what our opinion was. And then they could do the upgrade, so I think people were way happier. We got very few complaints, whereas we got a significant number of complains going from Hype 1.0 to 2.0.

So it’s really been kind of a win-win situation for our bottom line, for the future development of Hype and for the users.

Alexis: Alright. I could go on for about another hour and a half or longer, asking questions. But I think it’s [chuckles] it’s close to time that I start wrapping it up.

So, what’s one mistake you’d rather not repeat?

Jonathan: Oh, that is a good question. You know, let me talk because it’s fresh in my mind. I don’t think this is our biggest mistake, but this is one thing early on.

Since I was talking about our support load, one thing we focused on a little bit too much was e-mail support, and I really wished we had focused more on forum and public support.

Alexis: Interesting.

Jonathan: So we recently switched to a form of software called Discourse which did not exist back then, but we have our own forum. But the new Discourse forums, for whatever reason, just the way they’re structured, the way people can get replies; it’s really fostered a community of users that help each other.

We’ll still answer on the forums as well, and those are public answers so anytime we do a public answer, there a good chance we won’t get an inbound e-mail, which is a one-to-one communication, which is a one too many communication.

I think we just kind of said, “Well these are the people that are personally emailing us,” and I don’t know what we were thinking at that time. I think we felt like they wanted that one-to-one attention, and so we were giving them more attention than we probably should have via e-mail. But then having the right forum software would’ve been huge in those early days where we would’ve helped develop more of a community where there’d be more people figuring out the solutions and how to accomplish things with Hype.

So I really wished we had, from the get-go, we spent more time on the public, open discussions, and helped foster a larger community publicly. That would’ve saved our skins quite a bit in those beginning days where we were in such customer support hole.

Alexis: Do you still do e-mail support?

Jonathan: Yeah, we still do e-mail support. I love e-mailing with users, but our forums have just really taken flight. We did the new forums when we introduced Hype 3.0 and Hype Pro, and they’ve just really taken off.

It’s so fascinating to see – one of the new features we added was the ability to do physics-based animations, because sometimes doing key frame animations, it’s really hard to do certain effects, so we said, “Let’s try to take this to the next level and introduce physics into Hype.”

And so there’s this one thread of people trying to make physics-based games, and it was one person saying, “Oh this is a technique I found” and then the next post would be, “Oh, but you can make it better by doing this.”

And so you had a thread, I don’t know, it’s maybe like 50 or 60 posts of just people helping each other, taking this technology, which is really brand new technology in Hype, to a level that I had no clue it could ever go. I barely even interacted on the thread at all; it was just people in the community doing it.

So it was a wonderful experience to see that type of interaction between people just taking a product we made at this level that we could never have conceived.

Alexis: And what’s funny is I bet that that helped your brains marinate on future features.

Jonathan: Oh, absolutely. Definitely it does. So many of the features we do are based on the feedback from users and we kind of collect and we reprocess and re-think of it, but it’s easy to see what people want to do and what the really interesting things are, and we get excited by that.

Alexis: Alright. Last two questions here. On the flipside, what’s one decision that you’re pretty proud of? It’s dangerous asking those two questions together because they can often be the flipside of the other one.

Jonathan: One thing that we did – and this isn’t what I’m maybe most proud of – but I’m sitting in our office right now, so I’m happy to say I’m very proud of the office that we have and not just like what it looks like.

When we started to grow a little bit, we found a space in Hayes Valley that was much bigger than what our company size would be. So we went in with a couple other companies to share the space, and I think that’s a really good decision to have because our company was small.

At the time, we were just four people and we would’ve been the same four people seeing each other day in and day out. So we said, “We’ll get this bigger office space, we’ll sublet it to people that we like or we find interesting, or have interest in businesses.”

And so now we’ve got over a dozen people in the office coming and going, doing very interesting projects. Some people are, of course, doing programming projects, but some people are doing video directions, some people are doing special effects.

We’re all able to feed off our energy, and because some of the fields are slightly different, we’re able to kind of get different perspectives and help each other out, like when we have needs to do videos, like some of the guys will help us out, and when they have needs for programming, we can help them out. So it’s just created a really nice culture that’s not just our company and the same four faces; it’s more people. That’s one thing I am proud of, that we’ve kind of helped build this office environment.

That’s not everyone working for Tumult unfortunately, but it is just a really great work environment that we can come into and share and enjoy coming into the office every day.

Alexis: It’s pretty interesting, a co-working space throughout of Tumult. I never would have thought it.

Jonathan: Yep.

Alexis: Alright. So last question here and it’s probably the hardest of all: what’s your text editor of choice?

Jonathan: [Chuckles] For programming, it’s definitely Xcode. There’s no question about that. Most of what we do is in Objective-C. Even for the JavaScript portions, I use Xcode. I don’t feel it’s necessarily the best, but Xcode to me kind of reflects the general Macintosh key bindings as far as – I’m really fast with it. I can do all kinds of combinations with option and control and shift and the arrow keys, and move throughout the text pretty effectively.

It also, for a while, treated Monocode 9.0 very well. It doesn’t so much anymore, especially with the retina displays, but that’s okay. So yeah, I’m big into Xcode. I’ll use Emacs from time to time when I’m in the terminal, that felt the most Mac-like to me at that time when I was in college and I had that choice between Emacs and VI. I think there was the philosophy of ‘no modes’ which came from Jef Raskin and so I kind of took that to heart, and didn’t seem like Emacs had the same modes that VI had, so it seemed, on a philosophical level, Emacs was a little bit better. But most of the time I’m in Xcode, or if I’m doing pure text documents, I use Text Edit.

Alexis: Okay, no nvALT? That’s become my Bible as of late for note-taking and everything.

Jonathan: No, I just use text edit and just save things to Dropbox. I could be more organized, but I’m not. But I think there’s probably entrusting things people could do with note-taking as well.

Alexis: Alright. So if folks would like to learn more about Tumult or Hype, where should they go?

Jonathan: They should go to tumult.com or tumult.com/hype – that’s the best way to learn about it. We have all kinds of tutorials; we have gallery documents. There is a free trial on our website, which is really the best way to learn about Hype. Get it. We try to be as easy to use as possible. If you can use Keynote, you can pretty much use Hype.

Alexis: And if we’re curious about what you’ve had for lunch or what’s going on at the coworking space, where can we find you on Twitter?

Jonathan: So my Twitter handle is just @jmfd.

Alexis: Alright. Jonathan, thanks for coming on the show and spending an hour with us.

Jonathan: Glad to be here.

Alexis: And I will have to talk to you again sometime, see if we can see you for another hour so I can get to the second half of my questions.

Jonathan: Yeah, I would be happy to come back. Sorry if I need to elaborate quite a bit on those.

Alexis: No, no worries. They’re fantastic answers – thoughtful and thorough. Alright, as for the listeners, we’ll see you all the week after next.

The post Binpress Podcast Episode 36: Jonathan Deutsch of Tumult appeared first on Binpress.